Tool use is a defining feature of human culture, but the developmental origins of this ability remain controversial. One widely accepted view is that tool use represents a discontinuous developmental advance, dependent on the attainment of new symbolic or relational reasoning abilities. In contrast, this project examines the possibility that tool use may emerge more continuously during development. According to this view, tool use is rooted in the perception-action routines that infants employ to explore and act on their environments. In this connection, the present proposal examines how tool use may constitute an extension of infants'object manipulation skills. Three primary issues are addressed. First, the proposed studies consider how tool use may build on infants'efforts to relate objects to substrates in their surroundings. Second, the project explores whether tool use may be an outgrowth of infant object manipulation behaviors. For this purpose, longitudinal studies are proposed to gather kinematic data on object manipulation and tool use behaviors that are hypothesized to involve similar manual actions. Third, the proposed research examines how infants may learn about potential relations between objects and substrates by observing the outcomes of actions performed by others. Across the project, these issues will be addressed in a series of experiments with infants from 6 to 15 months of age. The methodologies will involve behavioral observation and kinematic studies. Taken together, the results of the proposed work are likely to illuminate how tool use develops in infants, suggest interventions for promoting tool use in typically and non-typically developing children, and inform debate over the extent to which tool use is uniquely human.